ABSTRACT / PROJECT SUMMARY Limited English proficiency (LEP) is a unique vulnerability of older immigrants that poses a significant risk to health and healthcare. Given that social and environmental contexts play a critical role in the lives of persons with LEP, this proposal investigates how social connectedness and neighborhood/community characteristics (e.g., ethnic density, health service environments in the neighborhood) influence the link between LEP and health/healthcare. We used Korean American elders as the target group. Our selection was based on the fact that they are members of a rapidly expanding LEP population (i.e., Korean is ranked 4th in the languages spoken by LEP individuals in the U.S.) and that they manifest marked disparities in health and healthcare. The project aims to explore the direct and interactive roles of social connectedness and neighborhood characteristics (1) in the relation between LEP and health and (2) in the relation between LEP and healthcare. The negative impact of LEP on health/healthcare is expected to be greater when individuals lack social connections and/or when they live in areas lack ethnically oriented resources and services. The project employs an innovative and synergistic mix of Social Network Analysis (SNA) and Geographic Information Systems (GIS). To capture the heterogeneity of the population, we will use 3 sites that combine to represent a continuum of Korean American population density: New York (high), Texas (intermediate) and Florida (low). Following focus groups with Community Advisory Boards, direct assessments of ethnic communities will be conducted. At each site, a master list of ethnically oriented services (including health services) will be generated, and each identified service will be geo-coded. Subsequently, 900 Korean Americans (aged ? 65, 300 at each site) will be surveyed using a probability sampling procedure. In addition to the traditional survey measures, name-generator approaches will be used to explore participants' social network members and places for healthcare. Using SNA, information from the name generators (names and addresses of people and places) will be indexed to represent (1) social connectedness and (2) engagement with health services at the community. Using GIS, the survey data will be linked to a file combining the 2010 Census and the area resource map derived from the direct community assessment. The combined file will include variables such as proportions of Koreans in the neighborhood and general as well as ethnically-oriented area health services. Multi-Level Models (MLM) will be used to explore the research questions and hypotheses. The results will not only enhance our understanding of the mechanisms underlying LEP vulnerability but also identify individual- and community-level factors (both in people and places) that could be used in health planning and interventions. The overall approach and findings will inform how to develop effective interventions to reduce language barriers and ensure access to appropriate health services for diverse LEP populations.